SYNTHESIS NOTE
TopicsRoutersthis note

Can routing beat building one better model?

Does directing queries to specialized models via semantic clustering outperform investing in a single frontier model? This challenges whether model improvement or model selection drives performance gains.

Synthesis note · 2026-02-23 · sourced from Routers

Avengers-Pro demonstrates that routing queries to different models based on semantic clustering can exceed the performance of any individual model in the pool — including frontier models. The mechanism: embed incoming queries, cluster by semantic similarity, evaluate per-cluster model performance-efficiency scores, and route each query to the highest-scoring model for its cluster.

Three results establish the claim:

The earlier Avengers work made an even more striking claim: ten models of ~7B parameters each, with routing, surpassed GPT-4.1 and 4.5 across 15 datasets. This suggests the performance gain from optimal model selection can be comparable to the gap between model generations.

The architecture is lightweight: three operations at inference time (embedding, nearest-cluster lookup, score aggregation). The heavy work — fitting the clustering model and estimating per-cluster performance statistics — happens offline on a calibration set (70% for fitting, 30% for evaluation). This makes the approach deployable as a thin routing layer atop any model API ecosystem.

Since Can we allocate inference compute based on prompt difficulty?, Avengers-Pro adds a complementary optimization axis. Compute-optimal scaling asks "how much inference budget per query?" Routing asks "which model per query?" These are independent — a routing layer could be composed with per-query compute allocation for a two-dimensional Pareto optimization. Since Can inference compute replace scaling up model size?, routing extends this: you don't need a bigger model OR more compute — you need the right model for this specific query type.

The implication challenges the frontier model race: rather than building one model that dominates on everything, assembling a diverse pool of specialized-ish models with good routing may be both cheaper and more effective. This aligns with the heterogeneous architecture thesis in Can small language models handle most agent tasks? — routing makes the heterogeneous approach practical.

Inquiring lines that read this note 68

This note is a source for these research framings, grouped by the broader line of inquiry each explores. Scan the bold lines of inquiry; follow any specific question forward.

How can LLM recommenders match or exceed collaborative filtering performance? Can model routing outperform monolithic scaling as an efficiency strategy? What structural factors drive popularity bias in recommendation systems? How does reasoning graph topology affect breakthrough insights and generalization? Can ensemble evaluation methods reduce bias more than single judges? Why do self-improving systems struggle without clear external performance metrics? What structural advantages do diffusion language models offer over autoregressive methods? Does domain specialization cause models to lose capabilities elsewhere? Does fine-tuning modify underlying model capabilities or only behavioral outputs? How do multi-agent systems achieve genuine cooperation and reasoning? How does example difficulty affect learning efficiency in language models? How do knowledge injection methods compare across cost and effectiveness? What are the consequences of models training on synthetic data? How should retrieval systems optimize for multi-step reasoning during inference? When does architectural design matter more than raw model capacity? How do evaluation mechanisms prevent error accumulation in autonomous research systems? Why does self-revision increase model confidence while degrading accuracy? What dimensions of recommendation quality do standard metrics miss? Can graph structure and relationships fundamentally improve recommendation systems? What causes silent corruption to amplify through delegated workflows? How do standardized protocols improve coordination in multi-agent systems? How do aggregate reward models systematically exclude minority user preferences? What memory abstraction level best enables agent knowledge reuse? When does optimizing for quality undermine the value of diversity? Does decoupling planning from execution improve multi-step reasoning accuracy? What memory architectures best support persistent reasoning across extended interactions? What makes weaker teacher models effective for stronger student training? Can single-axis benchmarks accurately predict agent deployment success? How should iterative research systems allocate reasoning per search step? How does objective evolution guide discovery better than fixed planning? Do harness improvements transfer across model scales or memorize shortcuts? How should personalization be implemented to improve AI assistant effectiveness?

Related concepts in this collection 4

This note in its neighbourhood — explore the map, then jump to a related concept in the list below.

Concept map
12 direct connections · 115 in 2-hop network ·dense cluster Open in graph ↗

Click a node to walk · click center to open · click Open in graph to see this note in the full knowledge graph

your link semantically near linked from elsewhere

Related papers in this collection 8

Papers most semantically related to this note, ranked by cosine similarity in the embedding space.

Original note title

test-time model ensembling via embedding-cluster routing surpasses any individual frontier model — model selection is a stronger lever than model improvement